Artificial intelligence in HIV research: a structured review and task-oriented clinical framework.

This opening section outlines how the systemic-relational approach in psychotherapy has engaged with the evidence-based movement, a dialogue marked by the attempt to make complexity visible without reducing its meaning—an ongoing challenge shared by several psychotherapeutic traditions. Retracing the theme of what can legitimately be considered “evidence” in psychotherapy, it reflects on the contribution that text-mining methodologies may offer to this matter. The section also provides an overview of Semantic Polarities Theory, whose presentation is essential as it constitutes the theoretical framework that structures and guides the present research. The history of systemic family therapy mirrors the broader trajectory of psychotherapy's encounter with the evidence-based paradigm. When Hazelrigg et al. () reviewed outcome studies, they already pointed out that systemic interventions seemed promising but also remarked the limited evidence supporting their effectiveness. Since then, the field has moved forward: in a more recent work adopting a meta-perspective on psychotherapy research methods, Castonguay et al. () included systemic therapies among the four main theoretical orientations recognized as evidence-based treatments for a wide range of clinical problems. These approaches often benefit not only identified patients but all members of the family system across diverse conditions—from externalizing and eating disorders to substance use disorders, depression, and psychosis—and have been successfully applied across the life span (children, adolescents, adults) and in multiple settings involving couples, family, individuals, groups, and multifamily interventions (–). Moreover, cost-effectiveness analyses suggest that healthcare systems might benefit from integrating systemic therapies into their services (). In line with this, recent large-scale data from integrated youth mental health programs show that many clients present with moderate to high clinical and psychosocial complexity, pointing to the need for more holistic and multidisciplinary approaches in public healthcare (). Yet as Sprenkle () argued, the mechanisms of change remain under-explored: although systemic treatments are known to produce effects, relatively few scholars have identified which components of those interventions are the active ingredients and which are inert fillers in specific approaches, so that the “Why” it works remains uncovered. In fact, in a comparative study classifying couples and family therapy models (), the authors note that not all systemic models share the same level of empirical backing: according to the guidelines of Sexton et al. (), some models are ranked as reaching only Level I (i.e., showing some empirical support but with methodological limitations), rather than fully evidence-based, that would be a Level III. For instance, the Milan systemic approach (–)—well known and widely recognized in Italy and internationally—has been categorized as a level I, therefore not fully evidence-based. Nevertheless, several efforts have been made to reconcile the systemic epistemology with empirical standards. Calls for manualization of systemic family therapy date back more than two decades () and the demand for a coherent, integrated assessment battery for outcome variables has persisted in subsequent years (). Notably, during those years there was a strong impetus to measure systemic therapy and several tools have been validated (, ) yet these instruments have not achieved widespread adoption, despite the broader movement toward empirically supported therapies.